Happy Thursday! Glad to announce new v2.10 release. It was released almost 2 weeks ago, but GooglePlay held update in hostages until I found a way to reach a person in google store team to help with that so I didn’t post changelogs until all platforms were deployed.
For those who don’t know - JellyBox is a music only opensource native crossplatform player for Jellyfin, Emby and Navidrome which I work on since 2023. Unlike other clients I focus on convenience, have spotify-like features. App runs on desktop, mobile and cars(android auto/carplay).
I also tried this new grid type of release notes(insprired by apple’s presentations), what do you think?
New Features
- Added mobile home-screen widgets(ios/android).
- Added Chromecast as a playback target. iOS might be flaky, sometimes it needs few times to cast and stop until it works. Not sure whats the problem yet.
- Added relay. Using relay you can connect to your jellyfin server remotely without need to use tailscale, VPN or port forwarding. Connect in settings. Needs you to install Jellyfin plugin.
- View lyrics in full-screen Studio mode, with word-by-word highlighting when available.
- Added ratings for Navidrome songs.
- Added a right side ruler for quickly navigating alphabetically sorted lists(enable in settings).
- Added support for custom ListenBrainz server addresses.
- Added a setting to clear cached images.
- Added a remote server address. This could be used for tailscale, VPNs or anything where your remote address differers from local.
Improvements
- Album pages have a refreshed design with animated cover art when available.
- Artists on album pages are now clickable.
- Albums on artist pages are sorted from newest to oldest and show their release year.
- Liked songs can be downloaded and, on Navidrome, sorted by when they were liked.
- Downloaded playlists and liked songs are now available in CarPlay and Android Auto. Queues can also be restored when the server is offline.
- Playback stays on the selected device when controlling another device, and reconnects more reliably after starting from a car or Bluetooth control.
- Switching back from AirPlay is clearer: the current device remains hidden until playback is switched back.
- Desktop window controls are now available throughout the app, including the login screen.
- Studio mode windows can now be dragged and resized on Windows and Linux.
- Horizontal rows no longer scroll unexpectedly when using a mouse wheel.
- Home-page rows support pull-to-refresh(from left to right).
- Control colors better match the current cover art, with an option to disable this behavior.
- Playlist cover art is now centered.
Bug Fixes
- Improved stability on Android, in-car use, and session restoration across devices.
- Fixed artist results from Navidrome being shown outside the selected library.
- Fixed playback queue items not being replaced correctly when starting the app or switching devices.
- Fixed the Linux app name appearing incorrectly in some places.
I’m still dealing with issue when you connect to bluetooth headunit in car with all that needs cold boot - sometimes playback not starts automatically. I’m aware of that and trying to fix it using my own car’s headunit. I’m not totally understand the issue yet, but seem like I’m not alone with it, other players have it too. Hope to find solution by next release.
https://github.com/avdept/JellyBoxPlayer/releases/tag/v2.10.0
PS: This is my first post on lemmy, I used to post changelogs only on reddit, but from now on - I will post them here too.
You must log in or register to comment.
Thanks! Didn’t know this existed and thanks for sharing. I have been mainly using AvSub for music playback with a Navidrome backend mainly because it has the ability to download music offline for playback on an Apple Watch.
JellyBox supports navidrome too, along with downloads too. And its free ofc
i never heard of jellybox before, how does it compare to plex?
jellybox is a client app, not server it works with your jellyfin, emby or navidrome server and content which you have there
That explains why I kept getting results for jellyfin when I googled it. Lol
Thanks for the course correction. I’ve barely started dipping my toe in this since my last attempt w mythbuntu and Kodi and jellyfin looks like a better fit.
This is really great! I might start self-hosting my favorite songs tbh
Looks interesting, always on the lookout for Android compatible Jellyfin music players with offline support. Maybe I’m mistaken but I think(?) last time I had looked at JellyBox it did not have offline support, nice to see it was added at some point.
Also, looks like there is already a Github issue about it but I was going to suggest having your app added to F-Droid if possible.
PS - Welcome to Lemmy!
its have downloads since like 4-5 months I think
about fdroid - thats on my list for sure
and thanks for welcome, happy to be here!
+1 for the F-Droid listing, instantly makes an app a lot more trustworthy and install-worthy to me
This looks a great bridge to move on from Plex for people like me who also rely on PlexAmp.
I rely on PlexAmp, been hoping to switch to Jellyfin though. Does JellyFin not have a native music solution at all? Or is this just a fancier version?
Jellyfin needs to handle all kinds of media. I tend to use Finamp for music and Findroid for TV/movies. Both support offline play, which last I checked Jellfin’s app still doesn’t.
I tried a few before settling on those, but I hadn’t heard of Jellybox until now so not sure how it compares.
I haven’t yet built a Jellyfin docker so I’m not sure, but I understand it’s like regular Plex - functional but not really built as a music-first player. I view this as pretty much the same as PlexAmp - accessing the Jellyfin database but giving way more features and options for music. So things like this really help.
Totally! I spend a lot of time on various convenience features to make UX as smooth as possible
This looks nice so far! Working well on Windows. Will be trying it on Android & Linux too.
Would be nice if it integrated with the Windows OS media interface or whatever it’s called (unless I’m missing an option?).
Hey, thanks
Windows has media interface widget integration(the one you open clicking on sound icon in taskbar bottom right corner)
Oh yeah, that does work, I guess I just forgot where it actually went, haha. Thanks for the quick response!
you’re welcome! feel free to ask if anything not clear
im about to go to bed for today, but will check and reply tomorrow
Trying it out right now, and I’m already impressed.
happy to hear! You’ll like all the hidden convenience things that will pop up as you use it more
One thing I’ve noticed is it’s a little buggy in CarPlay. It will work when you play something from the phone, but selecting music from the car screen doesn’t work. Also it won’t always start back up when you get back in the car.
Please keep working on it, cause it’s very nice to use.
Yes, I’m aware of these issues on carplay and working on resolving it!
Thanks. Could you provide an answer the latter question?
So you’re saying that everything, aside from the README is human-written with no LLM output?
README smells like slop. Are you using LLMs for this or no? All the telltale signs are there.
Edit: I’m going to take OP’s avoidance of my direct question about the code being written by LLMs as a “yes”. Disappointing.
I develop app since 2023, even before LLMs were a thing, so no, its not vibe coded. Although I used LLM to rewrite readme, original one was super terrible
If that’s the case, it might be good to rewrite the README as a human. Anyone that despises LLMs is going to see that and draw the same conclusion I did.
So you’re saying that everything, aside from the README is human-written with no LLM output? If so, that’s really cool, and I look forward to giving it a try!
tbh it would be hard - i’m really bad at writing readmes, and I spend most of the time out of my 9-5 and family time to add new features and readme is probably not the most important feature.
however I would gladly accept any help if anyone willing to do it
Thanks. Could you answer the latter question as well?
So you’re saying that everything, aside from the README is human-written with no LLM output?
